AI Factory Network Architect (AI)

Location: Remote within the United States or United Kingdom; travel to pod sites in the US and Europe is required for bring-up and major changes.

Company

hirify.global builds and operates an AI inference platform and its own distributed AI infrastructure, including Sonic Inference Pods with custom-designed compute and large-scale GPU fabrics.

What you will do

• Own the end-to-end AI Factory network architecture across NVLink/NVSwitch scale-up fabrics, InfiniBand and RoCE scale-out fabrics, storage and management networks, and pod-to-pod and inter-site backbones.
• Design the optical layer, including 400G/800G transceivers and cabling, co-packaged and linear-drive optics, DWDM, coherent optics, link budgets, and reliability targets.
• Define a validated reference network design for Sonic Inference Pods and build the first fabrics, including rail-optimized topologies, subnet management, routing, adaptive routing, congestion control, QoS, and GPUDirect RDMA tuning.
• Align topology, buffering, and network decisions with inference traffic such as tensor- and pipeline-parallel collectives, KV-cache transfer, and multi-node serving.
• Lead capacity planning, vendor roadmaps, technical specifications, fabric-generation evaluations, telemetry, link-health monitoring, failure-domain design, automated fault isolation, and operational runbooks.
• Hire and lead network engineers and serve as the final technical escalation for fabric incidents across sites.

Requirements

• 15+ years designing, building, and operating high-performance networks, including significant experience with HPC or AI supercomputers at the scale of thousands of GPUs or more.
• Deep hands-on InfiniBand experience covering fabric design, subnet management, routing engines, adaptive routing, congestion control, and port- and queue-pair-level performance debugging.
• Working knowledge of NVLink and NVSwitch domains and the interaction between scale-up and scale-out fabrics for multi-GPU and multi-node inference and training.
• Strong optical networking experience with 400G and 800G transceivers and cabling, link budgets, DWDM, coherent optics, and large-scale optics operations.
• Hands-on RDMA experience across InfiniBand and RoCE, including GPUDirect, NCCL, collective behavior, PFC, ECN tuning, and their trade-offs.
• Experience leading network teams through greenfield builds and production incidents, including complex fabric troubleshooting.
• Ability to automate infrastructure with Python or Go, infrastructure as code, and telemetry pipelines rather than CLI-only workflows.
• Clear technical writing and ability to turn designs into executable documentation for site teams.
• Willingness to travel to pod sites in the US and Europe for bring-up and major changes.

Nice to have

• Experience with NVIDIA Spectrum-X, Quantum-X800, Blackwell-generation NVLink systems, or equivalent hyperscaler technology.
• Experience with co-packaged optics, linear-drive pluggables, or photonic switching evaluations.
• Experience designing networks for modular or containerized data centers, edge sites, or deployments with constrained power and space.
• Experience with inference-specific traffic patterns, including disaggregated prefill and decode, KV-cache transfer, and multi-tenant serving.
• Contributions to SONiC, OpenFabrics, UEC, or similar open networking projects.
• Experience with sovereign or data-residency network designs, inter-site encryption, and segmentation.
• Prior work at a GPU cloud, hyperscaler, national lab, or AI lab building its own supercomputer.

Culture & Benefits

• Remote-first work with in-person company gatherings twice a year.
• Core collaboration hours with flexible scheduling outside those blocks.
• Fast, intense release cycles followed by dedicated downtime.
• Paid vacation, sick days, public holidays, and family leave.
• Meaningful stock options and twice-yearly company retreats.